Sleep Train lets you write scripts that will be played back to you while you sleep. You can customize your scripts exactly how you want, saying whatever you want. There are no audio packs to buy, no pre-scripted content—everything is customizable.
Sleep Train uses an advanced, on-device language model, Kokoro, that provides better text-to-speech than Apple’s built-in speech. The Kokoro model requires a recent iPhone running iOS 26+.
The Sleep Train app has three main pages: Scripts, Sessions, and Settings.
Each Script is a collection of lines. Each line is one spoken utterance. Tapping the name of a script navigates to the Script Editor, where you can edit, add, and delete lines.


Each session is a collection of scripts.

When you add a script to a session, you can set a duration for that script. The session plays lines from the script at random for the duration you set, then moves on to the next script.
If a script has only a few lines but a long duration, the app repeats lines as needed to fill the time. It tries to play every line before repeating any, and will not play the same line twice in a row.
Sleep Train has a wide variety of audio settings to tune the voice exactly how you want it.

Time Between Clips sets the pause between each line when played back. A short duration reads lines quickly one after the other.
Fade In/Out provides a smooth fade at the start and end of a session—both for the background audio and for your first and last spoken clip.
Background Audio is a Pro Audio feature and requires purchase of the Pro Audio item. Choose from several audio effects that play in the background while your session runs. A separate Background Volume control lets you balance the background audio against the spoken script.

TTS Engine — By default, standard Apple Speech is used, and you can select any Apple voice. If you have iOS 26+, you can also select the Kokoro Engine, a separate on-device language model that provides higher-quality speech. The first time you use Kokoro it downloads the model to your phone, which may take a minute. All speech processing happens locally on your device—no part of your scripts is ever sent anywhere.
Voice — Select either an Apple or Kokoro voice for playback. Changing the voice or any voice-related parameter plays a short demo so you can hear what it sounds like.
Speech Rate — Adjusts the speaking tempo, faster or slower.
Voice Pitch — Raises or lowers the pitch of the voice.
Voice Volume — Usually set at 100%, but you can lower this to balance the voice against background audio.

Timbre Settings is another Pro Audio feature unlocked with the Pro Audio purchase. Background Audio and Timbre are bundled together—purchasing Pro Audio unlocks both.

Sleep Train renders the text of your scripts into audio as a background process, enabling instant playback without delay. The cache rebuilds automatically whenever you change a script or audio settings. If you notice any issues, you can manually rebuild the audio cache from the Settings page.
For the simplest operation, tap the Play Session button when you go to sleep. Sleep Train also integrates with iOS Shortcuts, so you can schedule sessions to start automatically.

Your phone will now play your sleep session at the scheduled time every night.
For more control, you can trigger the session after a delay from bedtime rather than at a fixed clock time. This uses a Hold focus.
Download this shortcut: Hold Timer
When run, the Hold Timer shortcut prompts you for a duration, turns on your Hold focus, then turns it off after that time. For example, if you set a 2-hour hold when you go to bed, the focus turns off 2 hours later.
This way, your session starts automatically after your chosen delay—no matter what time you go to bed.
For information about how Sleep Train handles your data, see the Privacy Policy.